Advertisement

Natural Hazards

, Volume 99, Issue 3, pp 1233–1257 | Cite as

Impact of urbanization on hydrological processes under different precipitation scenarios

  • Wenbin Zang
  • Shu Liu
  • Shifeng HuangEmail author
  • Jiren Li
  • Yicheng Fu
  • Yayong Sun
  • Jingwei Zheng
Original Paper
  • 418 Downloads

Abstract

According to analysing the trends of land use changes in the upper reaches of Minjiang River in the past 30 years and precipitation in the last 50 years, nine types of simulation scenarios were constructed for different precipitation conditions and urbanization development processes. Based on the “five sub-basin selection principles” and “two simulation results evaluation indicators” proposed, the paper studied the influence of the urbanization process on hydrological processes under different precipitation conditions using the SWAT model. The primary conclusions are as follows: (1) the simulation results under the two kinds of land use transfer scenarios show the same laws: (a) when forest land (or grassland) is transferred to urban land, actual evapotranspiration (ET), soil water content (SW), amount of water percolating out of root zone (PERC) and groundwater contribution to streamflow (GW_Q) show a decreasing trend, and the reduction in watershed hydrological indexes is manifested as “high precipitation > average precipitation > low precipitation”. Moreover, surface runoff (SURQ), water yield (WYLD) and annual runoff show an increasing trend, and the increment in SURQ shows “high precipitation > average precipitation > low precipitation”, while the increment in WYLD and the simulated annual runoff show “low precipitation > average precipitation > high precipitation”. (b) Through analysis of the contribution of unit proportion transfer (CUPT) of watershed hydrological indicators, “SURQ > PERC > GW_Q > ET > SW” is observed in all precipitation scenarios. (2) Comparing simulation results between the two kinds of land use transfer scenarios: the CUPT variations of ET, SURQ and WYLD and the contribution of unit area transfer variations of daily flood peak and annual runoff both show “forest land transfer to urban land > grassland transfer to urban land”. Finally, two special phenomena observed in the analysis of the simulation results were discussed. The study results can provide a scientific basis for urban planning and construction for reducing the impact on urban flood.

Keywords

Urbanization Land use change Precipitation scenario SWAT model 

1 Introduction

Since the twenty-first century, global extreme weather has become more frequent (Hirabayashi et al. 2013). Climate change has a profound impact on the Earth system (People’s Daily 2012; Lesk et al. 2016). In 2012, Intergovernmental Panel on Climate Change (IPCC) released a report on the merits and demerits of the most comprehensive assessment of the impact of climate change on floods and other extreme disasters (IPCC 2012). The assessment results indicate that extreme weather will lead to increased risk of flooding (IPCC 2012). The European Commission released a flood guideline to guide flood risk management in EU member states, which explicitly pointed out that the risk assessment must pay attention to the impact of climate change (European Commission 2007). The climate change in China shows a considerable similarity to the global change through the analysis of climate data in the past 100 years (Ding et al. 2007).

Human endeavours, including agriculture, forestry, fishing, animal husbandry, mining, industry, business, transportation, sightseeing and various construction projects, are a series of activities used by different types of people to survive and improve their living standards (Ye et al. 2001). Human beings affect the hydrological cycle by changing the river channel (Liu et al. 2014), changing the underlying surface (Zhao et al. 2013), excessive exploitation of groundwater (Kaushal et al. 2017) and other activities (Cornelissen et al. 2013; Ellis and Ramankutty 2008). Urbanization is one of the most important factors in human activities affecting hydrological processes (Zhang 2012). The process of urbanization changes the characteristics of the underlying surface (Breuer et al. 2009; Zhang et al. 2014). In particular, the increase in impervious area brought by urbanization has greatly affected the process of runoff and concentration on urban surfaces (Song et al. 2014). Urbanization brings a range of environmental challenges as a direct result of the biochemical and physical changes to hydrological systems (Zheng et al. 2003; Fletcher et al. 2013).

Hydrological modelling plays a notably important role in assessing the impact of land use changes and climate changes on hydrological systems (Fletcher et al. 2013; Dwarakish and Ganasri 2015; Piao et al. 2007). Onstad and Jamieson (1970) proposed a hydrological model to study the impact of land use changes on runoff first. Since then, more and more hydrological models have been used in these studies, such as Soil and Water Assessment Tool model (SWAT) (Sajikumar and Remya 2015), Variable Infiltration Capacity model (VIC) (Hengade and Eldho 2016), TOPgraphy-based hydrological model (TOPMODEL) (Dietterick et al. 1999) and Xin’anjiang model (Xian et al. 2017). Many scholars (Sajikumar and Remya 2015; Jayakrishnan et al. 2005; Zhou et al. 2013) believe that the runoff simulation accuracy of SWAT is better. It is one of the more mature hydrological models in the world and one of the most widely used distributed hydrological models.

At present, scholars (He et al. 2013; Chu et al. 2013; Chen et al. 2015; Akhter and Hewa 2016) have examined the impact of urbanization on river peak flow and average runoff. However, the studies on hydrological effects of urbanization seldom take climate change into account, nor the horizontal comparison between urbanization scenarios. In this paper, the SWAT model will be used to study quantitatively the impact of urbanization on watershed and river hydrological processes under different precipitation conditions. The different hydrological responses of different types of urbanization will also be analysed. In this paper, “five sub-basin selection principles” and “two simulation results evaluation indicators” are proposed, which will help other scholars carry out similar research.

2 Materials and methods

2.1 Study site

The upstream of the Minjiang River, primarily from the source of Minjiang River to Zipingpu Reservoir, was chosen as the whole study area. The area includes the counties of Wenchuan, Li, Mao, Heishui and Songpan in Aba Tibetan and Qiang autonomous prefecture, Sichuan province, China.

The study area is located in northern Sichuan province, eastern Aba autonomous prefecture and in the Qinghai-Tibet Plateau region. Beichuan, An and Mianzhu counties are to the east of the study area, Chongqing and Dayi counties are to the south, Hongyuan and Maerkang counties are to the west, and Jiuzhaigou and Ruoergai counties are to the north. The area of the basin is approximately 230 thousand km2. The longitude is from 102°36′E to 103°58′E, and the latitude is from 30°46′N to 33°09′N. The location of the study area is shown in Fig. 1.
Fig. 1

Map of the upstream of Minjiang River

2.2 Materials

Daily precipitation data for 1977–1980 from seventeen rainfall stations, about 50 years of daily weather data (precipitation, maximum temperature, minimum temperature, average wind speed, relative humidity and solar radiation) from five weather stations (Wenchuan, Li, Mao, Heishui and Songpan) and daily runoff data for 1977–1980 from seven hydrological sites (Zipingpu, Shengliba, Zhenjiangguan, Heishui, Shaba, Zagunao and Shouxi) were collected in the paper.

DEM data were collected from National Aeronautics and Space Administration (NASA) with a spatial resolution of 30 metres (Fig. 2), and soil data from Institute of Soil Science, Chinese Academy of Sciences (1:1 million, shapefile data). In addition, the paper also obtained land use data with a spatial resolution of 1 km in 1980, 1990, 1995, 2000, 2005, 2010 and 2015 [raster data, the dataset was provided by Data Center for Resources and Environmental Sciences, Chinese Academy of Sciences (RESDC) (http://www.resdc.cn)] (Fig. 3).
Fig. 2

The spatial distribution information of terrain

Fig. 3

The spatial distribution information of land use. a 1980; b 1990; c 1995; d 2000; e 2005; f 2010; g 2015

2.3 Methodology

The SWAT model (Neitsch et al. 2011) is a distributed watershed hydrological model developed by the US Department of Agriculture’s Agricultural Research Agency (USDA-ARS) in the 1990s. The model has a strong physical basis, and it is suitable for complex river basins with different soil types, land use patterns and management conditions. The model has a wide range of research. Scholars have conducted many studies by utilizing this model, primarily for such applications as exploring and improving the model, uncertainty research of the hydrological model, hydrological scale research, lack of data hydrological simulation, climate change simulation, land use/vegetation cover change simulation, water resource evaluation and management, non-point source pollution simulation, sediment simulation and agricultural management.

Based on an analysis of the land use change trend in the upper reaches of Minjiang River from 1980 to 2015, land transfer scenarios in the process of urbanization and precipitation scenarios were constructed. Next, this paper studied the impact of urbanization processes on hydrological processes under different precipitation scenarios with the SWAT model.

The relative error of flood total volume (Dv), the Nash–Suttcliffe coefficient (Ens) (Nash and Sutcliffe 1970) and the coefficient of correlation (R2) (Nagelkerke 1991) were chosen as goodness-of-fit indices to evaluate the performance of the SWAT model, as shown in Eqs. (1), (2) and (3):
$$ D_{\text{v}} (\% ) = \frac{100(P - O)}{O} $$
(1)
$$ E_{\text{ns}} = 1 - \frac{{\sum\nolimits_{i = 1}^{n} {(O_{i} - P_{i} )^{2} } }}{{\sum\nolimits_{i = 1}^{n} {(O_{i} - \bar{O})^{2} } }} $$
(2)
$$ R^{2} = \left[ {\frac{{\sum\nolimits_{i = 1}^{n} {\left( {O_{i} - \bar{O}} \right)\left( {P_{i} - \bar{P}} \right)} }}{{\sqrt {\sum\nolimits_{i = 1}^{n} {\left( {O_{i} - \bar{O}} \right)^{2} } } \sqrt {\sum\nolimits_{i = 1}^{n} {\left( {P_{i} - \bar{P}} \right)^{2} } } }}} \right]^{2} $$
(3)
where O and P represent measured flood total volume and calculated flood total volume, respectively; Oi and Pi represent measured flood peak and calculated flood peak, respectively when the data sequence is i; \( \overline{O} \) and \( \overline{P} \) are the mean value of measured flood peak and mean value of calculated flood peak, respectively; n is the length of data sequence.

2.4 Trend analysis of land use changes in the last 30 years

Land use data from 1980, 1990, 1995, 2000, 2005, 2010 and 2015 were analysed statistically. The information on land use area in the upper reaches of Minjiang River from 1980 to 2015 is shown in Table 1. According to Table 1, the area of forest land and grassland accounted for more than 97% of the upper reaches of Minjiang River from 1980 to 2015.
Table 1

Statistical information of land use area in the upstream of Minjiang River from 1980 to 2015; unit: km2

Land use

Year

Code

Name

1980

1990

1995

2000

2005

2010

2015

1

Cultivated land

595

529

518

608

556

547

545

 112

Hilly paddy field

0

3

1

3

3

0

0

 113

Plain paddy field

4

1

2

1

1

1

1

 121

Upland dry land

591

505

492

584

534

528

544

 122

Hilly arid land

0

8

11

4

2

2

0

 123

Plain dry land

0

2

3

4

4

4

0

 124

>25° sloping land

0

10

9

12

12

12

0

2

Forest land

10,237

10,489

10,424

10,228

9996

10,006

9998

 21

Woodland

3939

4625

4555

3943

3745

3756

3761

 22

Shrubland

5063

5352

5369

5034

4932

4922

4909

 23

Sparse woodland

1194

484

482

1210

1269

1277

1282

 24

Other woodland

41

28

18

41

50

51

46

3

Grassland

11,728

11,552

11,623

11,726

11,995

11,979

11,968

 31

High coverage grassland

2832

1553

2223

2840

2935

2942

2933

 32

Middle coverage grassland

8488

9548

8967

8477

8600

8579

8577

 33

Low coverage grassland

408

451

433

409

460

458

458

4

Water area

45

38

42

43

42

54

62

 41

River

3

4

1

3

3

3

3

 42

Lake

18

12

14

16

16

16

16

 43

Reservoir

0

0

0

0

0

12

20

 44

Permanent glacier snow

24

22

27

24

23

23

23

5

Construction land

12

9

12

13

19

19

32

 51

Urban land

3

2

4

4

6

6

6

 52

Rural residential land

9

7

8

9

13

12

15

 53

Other construction land

0

0

0

0

0

1

11

6

Unused land

10

10

8

9

19

22

22

 64

Wetlands

9

10

8

9

10

10

10

 65

Bare land

1

0

0

0

2

3

3

 66

Bare rocky texture

0

0

0

0

7

9

9

Total

22,627

22,627

22,627

22,627

22,627

22,627

22,627

2.4.1 Analysis of construction land change tendency

The change trend of construction land area in the upper reaches of Minjiang River from 1980 to 2015 is shown in Fig. 4. Overall, the area of construction land declined from 1980 to 1990, and 1990–2015 showed a gradual upward trend with a dramatic increase in 2010–2015. Specific to the subcategories: (1) the change trend of urban land was the same as that of construction land. (2) The change trend of area of rural residential land was the same as that of construction land from 1980 to 2015, and from 2005 to 2010. (Analysis shows the area reduction of rural residential land may be affected by the 2008 Wenchuan earthquake.) (3) There was no monitoring of other construction land in 1980–2005, while the area increased rapidly in 2010–2015.
Fig. 4

Tendency of construction land area in the upper reaches of Minjiang River from 1980 to 2015

2.4.2 Analysis of Woodland Change Tendency

The forest land in the upper reaches of the Minjiang River was dominated by shrubland and woodland. The change trend of forest land area in the upper reaches of Minjiang River from 1980 to 2015 is shown in Fig. 5. In general, the area of forest land increased from 1980 to 1990 and gradually decreased from 1990 to 2005, while it was basically stable in 2005–2015. Specific to the subcategories: (1) the area change trends of woodland and shrubland were basically the same as that of forest land. (2) The area of sparse woodland decreased in 1980–1990, gradually increased in 1995–2005 and remained stable in 1990–1995 and 2005–2015. (3) The area of other woodland accounts for less than 0.5% of forest land area, and its area change was very small from 1980 to 2015. This change had little effect on the overall change of forest land.
Fig. 5

Tendency of Woodland area in the upper reaches of Minjiang River from 1980 to 2015

2.4.3 Analysis of grassland change tendency

The area of middle coverage grassland was dominant with more than 70% of the grassland area in the upper reaches of the Minjiang River. The tendency of grassland area in the upper reaches of Minjiang River from 1980 to 2015 is shown in Fig. 6. In general, the area of grassland decreased from 1980 to 1990, increased substantially from 1990 to 2005, and remained stable in 2005–2015. Specific to the subcategories: (1) the area of high coverage grassland declined by nearly half in 1980–1990, basically increased in 1990–2005 and remained stable in 2005–2015. (2) The area of middle coverage grassland increased remarkably in 1980–1990, decreased in 1990–2000, increased in 2000–2005 and remained stable in 2005–2015. (3) The area of low coverage grassland accounts for less than 5% of the grassland area, and the area was basically unchanged from 1980 to 2015.
Fig. 6

Tendency of grassland area in the upper reaches of Minjiang River from 1980 to 2015

2.5 Construction of scenarios

Based on an analysis of land use area change trends in the upper reaches of the Minjiang River in the last 30 years, the paper studied the influence of the urbanization process on hydrological processes under different precipitation conditions. Considering that forest land and grassland occupied almost all of the area of the upper reaches of Minjiang River, it is more meaningful to study the effect of urbanization of forest land or grassland transfer to urban land on the impact of hydrological processes. The paper constructed nine types of research scenarios to study quantitatively the impact of urbanization on watershed and river hydrological processes under different precipitation conditions (Table 2).
Table 2

Nine research scenarios

Land transfer scenario

Precipitation scenarios

High precipitation

Average precipitation

Low precipitation

Original land use

Scenario 1

Scenario 2

Scenario 3

Forest land transfer to urban land

Scenario 4

Scenario 5

Scenario 6

Grassland transfer to urban land

Scenario 7

Scenario 8

Scenario 9

According to research results in the literature (Huang et al. 2015), the precipitation of 1977–1980 belongs to a plain water year in the upper reaches of the Minjiang River. In this paper, three precipitation scenarios were established. Observed precipitation in 1977–1980 was suggested as the baseline scenario, while high precipitation scenario was increased by 20%, and low precipitation scenario was reduced by 20%.

2.6 Research indicators

Combined with the two concepts of contribution of unit transfer proposed in this paper, hydrological indicators of watershed and river were used for quantitative analysis of the influence of different precipitation scenarios of city urbanization on hydrological processes.

2.6.1 Watershed hydrological indicators

The main watershed hydrographic indicators were used in the study as follows (Arnold et al. 2011):
  1. 1.

    Actual evapotranspiration (ET): Actual evapotranspiration from the sub-basin during the time step (mm).

     
  2. 2.

    Soil water content (SW): Amount of water in the soil profile at the end of the time period (mm).

     
  3. 3.

    Amount of water percolating out of root zone (PERC): Water that percolates past the root zone during the time step (mm). There is potentially a lag between the time the water leaves the bottom of the root zone and reaches the shallow aquifer. Over a long period of time, this variable should equal groundwater percolation.

     
  4. 4.

    Surface runoff (SURQ): Surface runoff contribution to streamflow during time step (mm H2O).

     
  5. 5.

    Groundwater contribution to streamflow (GW_Q): Water from the shallow aquifer that returns to the reach during the time step (mm).

     
  6. 6.

    Water yield (WYLD): The net amount of water that leaves the sub-basin and contributes to streamflow in the reach during the time step (mm H2O).

     

2.6.2 River hydrological indicators

The main river hydrographic indicator used in this paper was as follows (Arnold et al. 2011):
  1. 1.

    Average daily streamflow into reach (FLOW_IN): Average daily streamflow into reach during time step (m3/s).

     

2.6.3 Contribution of unit transfer

  1. 1.
    The contribution of unit proportion transfer (CUPT): To facilitate quantitative description of the impact of land use change on watershed hydrological indicators, and the horizontal analysis of simulated results between the two types of land transfer scenarios, the contribution of unit proportion transfer is defined as the variation of ET, SW, PERC, SURQ, GW_Q and WYLD, while the land use for every 10% of the sub-basin is transferred to urban land. The formula of CUPT is as follows:
    $$ {\text{CUPT}} = \frac{{{\text{WHI}}_{1} - {\text{WHI}}_{0} }}{{10{\text{\% }}*S}}*S_{t} $$
    (4)
    where \( {\text{WHI}}_{0} \) is watershed hydrological indicators (ET, SW, PERC, SURQ, GW_Q and WYLD) of simulated results in original land use; \( {\text{WHI}}_{1} \) is watershed hydrological indicators (ET, SW, PERC, SURQ, GW_Q and WYLD) of simulated results when the land use of the sub-basin is transferred to urban land; \( S_{t} \) is the area of land use which is transferred to urban land; and \( S \) is the area of selected sub-basin.
     
  1. 2.
    The contribution of unit area transfer (CUAT): To facilitate quantitative description of the impact of land use change on river hydrological indicators, and the horizontal analysis of simulated results between the two types of land transfer scenarios, the contribution of unit area transfer is defined as the variation of FLOW_IN, while the land use for every 100 km2 of the sub-basin is transferred to urban land. The formula of CUAT is as follows:
    $$ {\text{CUAT}} = \frac{{{\text{RHI}}_{1} - {\text{RHI}}_{0} }}{100}*S_{t} $$
    (5)
    where \( {\text{RHI}}_{0} \) is river hydrological indicator (FLOW_IN) of simulated results in original land use; \( {\text{RHI}}_{1} \) is river hydrological indicator (FLOW_IN) of simulated results when the land use of the sub-basin is transferred to urban land; \( S_{t} \) is the area of land use which is transferred to urban land (km2).
     

3 Results and discussion

3.1 SWAT model construction and calibration

3.1.1 SWAT model construction

Based on the DEM data, land use data and soil data mentioned earlier, an underlying model of the upper reaches of Minjiang River was constructed. The study set 100 km2 as the threshold of catchment area in river network extraction, and the slope was divided into five grades according to 0–10%, 10–25%, 25–50%, 50–100%, > 100% (taking into account the suitable slope for urban land is 25% or less). The multiple hydrological response units (HRUs) way was selected in HRU division. The minimum threshold ratio of the soil type to slope type was 10%, and the minimum threshold ratio of land use type was 0% (that is, all land use types were considered in HRU division). The model divided 114 sub-basins (Fig. 7) and 2653 HRUs.
Fig. 7

The spatial distribution of hydrometeorological station and sub-basin division

Rainfall data (from seventeen rainfall stations and five weather stations) and temperature data (from five weather stations) were input as model weather conditions. Construction of a weather generator utilizing five weather stations from 1957 to 2008 was used to simulate the average wind speed, relative humidity, and solar radiation in the upper reaches of Minjiang River. The Skewed Normal method was used to simulate the spatial distribution of precipitation data. The Penman/Monteith model was used for evaporation simulation. The SCS model was selected for a runoff producing simulation. The Muskingum law was selected for concentration simulation.

3.1.2 Land use transfer scenarios in SWAT model

To facilitate the quantitative study of the impact of “forest land transfer to urban land” and “grassland transfer to urban land” in the urbanization process on watershed and river hydrological processes, this paper chooses a sub-basin according to the following principles (abbreviated as “five sub-basin selection principles”): (1) the selected sub-basin should contain the county town of 5 counties in the upper reaches of Minjiang River. (2) Referring to the literature (Sichuan Institute of Urban Planning & Design 1999), the greatest suitable slope for urban land is 25%. To facilitate horizontal comparative analysis between the two land transition scenarios, the proposed sub-basin with slope below 25% should has both forest land and grassland areas. (3) The soil types of HRUs in the sub-basin selected for “forest land transfer to urban land” or “grassland transfer to urban land” should belong to the same soil hydrological unit. (4) To ensure the accuracy of the quantitative analysis, the area of selected forest land HRUs or grassland HRUs should not be too small; the area should be more than 10 km2, and the proportion of the sub-basin should be greater than 1%. (5) The selected forest land HRUs or grassland HRUs should belong to the same slope level (0–10% or 10–25%).

According to the above principles, No. 84 sub-basin (Maoxian county seat) was selected as the research object (Fig. 8). The total area of No. 84 sub-basin was 68 401.0 ha, including 41 HRUs. The HRUs information is shown in Table 3. The construction of two urbanization process scenarios: (1) forest land transfer to urban land: No. 1904 HRU and No. 1914 HRU were transferred to No. 1928 HRU. Next, the area of No. 1928 HRU is 2 298.4 ha, accounting for sub-basin area of 3.36%. (2) Grassland transfer to urban land: No. 1919 HRU and No. 1926 HRU were transferred to No. 1928 HRU. Then, the area of No. 1928 HRU is 1 310.5 ha, accounting for sub-basin area of 1.92%.
Fig. 8

The comparison of SWAT model simulation and observed monthly discharge at Shengliba station

Table 3

List of HRUs in No. 84 sub-basin in the upper reaches of Minjiang River

Code

Lucc

Composition

(Lucc/soil/slope)

Area (ha)

Proportion (%)

Code

Lucc

Composition

(Lucc/soil/slope)

Area (ha)

Proportion (%)

(1)

(2)

(3)

(4)

(5) = (4)*100/68,401

(1)

(2)

(3)

(4)

(5) = (4)*100/68,401

1901

Forest-Mixed

FRST/anzongrang/50-100

3189.6

4.66

1922

Pasture

PAST/shihuixinghetu/25-50

3321.9

4.86

1902

 

FRST/anzongrang/25-50

1976.0

2.89

1923

 

PAST/zaohetu/25-50

1815.1

2.65

1903

 

FRST/heizhantu/25-50

2935.2

4.29

1924

 

PAST/zaohetu/50-100

2817.7

4.12

1904

 

FRST/heizhantu/10-25

1115.2

1.63

1925

 

PAST/zongrang/50-100

3679.2

5.38

1905

 

FRST/heizhantu/50-100

2973.1

4.35

1926

 

PAST/zongrang/10-25

838.5

1.23

1906

 

FRST/shihuixinghetu/50-100

6225.2

9.1

1927

 

PAST/zongrang/25-50

2923.9

4.27

1907

 

FRST/shihuixinghetu/25-50

3760.1

5.5

1928

Residential–high density

URHD/zaohetu/10-25

16.1

0.02

1908

 

FRST/zaohetu/0-10

888.3

1.3

1929

 

URHD/zaohetu/0-10

109.9

0.16

1909

 

FRST/zaohetu/10-25

1127.6

1.65

1930

Residential–med/low density

URML/shihuixinghetu/50-100

2.0

0.01

1910

 

FRST/zaohetu/25-50

1447.1

2.12

1931

 

URML/shihuixinghetu/25-50

5.0

0.01

1911

 

FRST/zaohetu/50-100

1424.9

2.08

1932

 

URML/shihuixinghetu/10-25

2.0

0.01

1912

 

FRST/zongrang/50-100

5028.9

7.35

1933

 

URML/zaohetu/25-50

15.2

0.02

1913

 

FRST/zongrang/25-50

3881.6

5.67

1934

 

URML/zaohetu/10-25

26.4

0.04

1914

 

FRST/zongrang/10-25

1183.1

1.73

1935

 

URML/zaohetu/0-10

26.4

0.04

1915

Pasture

PAST/anzongrang/25-50

1451.2

2.12

1936

Winter wheat

WWHT/shihuixinghetu/10-25

224.3

0.33

1916

 

PAST/anzongrang/50-100

2181.4

3.19

1937

 

WWHT/shihuixinghetu/50-100

438.3

0.64

1917

 

PAST/anzongrang/100-9999

479.3

0.7

1938

 

WWHT/shihuixinghetu/25-50

693.5

1.01

1918

 

PAST/heizhantu/50-100

1885.2

2.76

1939

 

WWHT/zaohetu/25-50

584.0

0.85

1919

 

PAST/heizhantu/10-25

472.0

0.69

1940

 

WWHT/zaohetu/10-25

259.1

0.38

1920

 

PAST/heizhantu/25-50

1521.6

2.22

1941

 

WWHT/zaohetu/50-100

391.9

0.57

1921

 

PAST/shihuixinghetu/50-100

5064.1

7.4

     

3.1.3 Model calibration and verification

The warm-up period of this SWAT model was 1977, the model calibration period was 1978–1979, and the validation period was 1980. The paper analysed the sensitivity of the parameters by SWAT-CUP tool. Next, the six most sensitive parameters (Cn2, Canmx, Gwqmn, Alpha_Bf, Revapmn and Esco) were selected to calibrate this SWAT model by Shuffled Complex Evolution (SCE-UA) method. The comparison of SWAT model simulation and observed monthly discharge at Shengliba station is shown in Fig. 8. The simulation results of the model calibration and validation period evaluated by the indicators ENS, R2 and |Dv| are shown in Table 4. Table 4 shows that the accuracy of the runoff simulation is very good. It proves the SWAT model can be applied in hydrological modelling of the upstream of Minjiang River.
Table 4

Simulation accuracy of month runoff

Simulation scenario

ENS

R2

Dv (%)

Zhenjiangguan

   

 Calibration period

0.67

0.90

0.5

 Validation period

0.86

0.95

1.57

Shengliba

   

 Calibration period

0.85

0.89

− 4.25

 Validation period

0.91

0.97

− 6.7

Heishui

   

 Calibration period

0.82

0.91

− 8.9

 Validation period

0.79

0.88

− 13.72

Shaba

   

 Calibration period

0.85

0.89

− 12.59

 Validation period

0.82

0.94

− 14.15

3.2 Simulation results analysis

3.2.1 Forest land transfer to urban land

Hydrological simulations of daily scale, monthly scale and annual scale were carried out under the three kinds of precipitation scenarios for “forest land transfer to urban land” based on the SWAT model in the upper reaches of the Minjiang River. The simulation results were statistically analysed; the responses of the main river hydrological indicators in “forest land transfer to urban land” under the three kinds of precipitation scenarios’ annual scale simulation are shown in Table 5, and the responses of the main river hydrological indicators in “forest land transfer to urban land” under the three kinds of precipitation scenarios’ daily scale, monthly scale and annual scale simulations are shown in Table 6.
Table 5

Responses of the main watershed hydrological indicators in “forest land transfer to urban land” under the three kinds of precipitation scenarios’ annual scale simulation; unit: mm

Item

ET

SW

PERC

SURQ

GW_Q

WYLD

High precipitation

      

 Original land use

352.882

65.197

124.417

6.860

103.140

237.370

 Forest land transfer to urban land

351.513

64.823

119.314

14.824

98.289

238.735

 Change valuea

− 1.369

− 0.374

− 5.103

7.964

− 4.851

1.365

 CUPT

− 4.074

− 1.113

− 15.187

23.702

− 14.438

4.063

 Trend

Decrease

Decrease

Decrease

Increase

Decrease

Increase

Average precipitation

      

 Original land use

331.042

56.563

77.886

3.441

59.117

159.211

 Forest land transfer to urban land

329.678

56.275

74.692

9.116

56.162

160.656

 Change value

− 1.363

− 0.288

− 3.193

5.675

− 2.954

1.445

 CUPT

− 4.058

− 0.858

− 9.504

16.889

− 8.793

4.301

 Trend

Decrease

Decrease

Decrease

Increase

Decrease

Increase

Low precipitation

      

 Original land use

295.931

46.431

41.031

1.267

25.000

95.041

 Forest land transfer to urban land

294.597

46.282

39.405

5.003

23.573

96.495

 Change value

− 1.333

− 0.149

− 1.626

3.736

− 1.427

1.454

 CUPT

− 3.968

− 0.444

− 4.838

11.118

− 4.248

4.326

 Trend

Decrease

Decrease

Decrease

Increase

Decrease

Increase

aChange value: Compared with original land use, watershed hydrological indicators change under “forest land transfer to urban land”; positive sign means an increase, negative means a reduction

Table 6

Response of the main river hydrological indicators in “forest land transfer to urban land” under the three kinds of precipitation scenarios’ daily scale, monthly scale and annual scale simulations; unit: m3/s

FLOW_IN

High precipitation

Average precipitation

Low precipitation

Original land use

Forest land transfer to urban land

Change value

CUAT

Original land use

Forest land transfer to urban land

Change value

CUAT

Original land use

Forest land transfer to urban land

Change value

CUAT

Daily simulated flood peak

1755

1757

2.000

8.702

1304

1306

2.000

8.702

903.1

904.7

1.600

6.961

Average runoff

            

 January

24.057

24.047

− 0.010

− 0.044

16.587

16.583

− 0.003

− 0.015

9.893

9.891

− 0.002

− 0.007

 February

20.470

20.477

0.007

0.029

14.003

14.007

0.003

0.015

8.581

8.582

0.001

0.004

 March

51.787

51.817

0.030

0.131

37.720

37.743

0.023

0.102

24.117

24.130

0.013

0.058

 April

126.133

126.200

0.067

0.290

93.477

93.553

0.077

0.334

62.227

62.290

0.063

0.276

 May

257.733

257.933

0.200

0.870

192.433

192.600

0.167

0.725

127.100

127.233

0.133

0.580

 June

391.167

391.333

0.167

0.725

296.000

296.100

0.100

0.435

199.200

199.300

0.100

0.435

 July

508.967

509.100

0.133

0.580

393.833

393.933

0.100

0.435

264.933

265.067

0.133

0.580

 August

379.167

379.100

− 0.067

− 0.290

293.600

293.567

− 0.033

− 0.145

193.467

193.467

0.000

0.000

 September

507.600

507.600

0.000

0.000

395.467

395.500

0.033

0.145

271.100

271.100

0.000

0.000

 October

356.300

356.167

− 0.133

− 0.580

278.033

277.933

− 0.100

− 0.435

189.567

189.500

− 0.067

− 0.290

 November

199.867

199.833

− 0.033

− 0.145

154.200

154.133

− 0.067

− 0.290

102.157

102.127

− 0.030

− 0.131

 December

86.210

86.183

− 0.027

− 0.116

61.273

61.263

− 0.010

− 0.044

36.980

36.977

− 0.003

− 0.015

Annual runoff

243.376

243.404

0.028

0.121

186.265

186.289

0.024

0.105

124.575

124.603

0.029

0.125

  1. (a)

    Watershed hydrological analysis

     
The results (Table 5) under the three kinds of precipitation scenarios show: (1) with increasing precipitation, the annual scale simulation results for ET, SW, PERC, SURQ, GW_Q and WYLD increase. (2) When forest land is transferred to urban land, ET, SW, PERC and GW_Q show a decreasing trend, and the reduction in watershed hydrological indexes is manifested as “high precipitation > average precipitation > low precipitation”. Meanwhile, SURQ and WYLD show an increasing trend, and the increment of SURQ shows “high precipitation > average precipitation > low precipitation”, while the increment of WYLD shows “low precipitation > average precipitation > high precipitation”. (3) The CUPT of the watershed hydrological index shows “SURQ > PERC > GW_Q > ET > SW” in the three kinds of precipitation scenarios.
  1. (b)

    River hydrological analysis

     

The results (Table 6) under three precipitation scenarios show the following: (1) with increasing precipitation, the simulation results for annual runoff, monthly runoff and daily flood peak increased. (2) Based on the annual scale model, the change trend of annual runoff amount shows “low precipitation > average precipitation > high precipitation” when forest land transfers to urban land. (3) Monthly scale simulation results show that the monthly runoff increases in February–July, decreases in October and January and remains stable in August and September. (4) The change trend of daily flood peak shows “high precipitation = average precipitation > low precipitation”.

3.2.2 Grassland transfer to urban land

In view of “grassland transfer to urban land”, the paper carried out research similar to the “forest land transfer to urban land”. The simulation results were statistically analysed, and the responses of the main watershed hydrological indicators in “grassland transfer to urban land” under the three kinds of precipitation scenarios’ annual scale simulation are shown in Table 7; the responses of the main river hydrological indicators in “grassland transfer to urban land” under the three types of precipitation scenarios’ daily scale, monthly scale and annual scale simulations are shown in Table 8.
Table 7

Response of the main watershed hydrological indicators in “grassland transfer to urban land” under the three kinds of precipitation scenarios’ annual scale simulation; unit: mm

Item

ET

SW

PERC

SURQ

GW_Q

WYLD

High precipitation

      

 Original land use

352.882

65.197

124.417

6.860

103.140

237.370

 Grassland transfer to urban land

352.165

64.988

121.541

11.279

100.408

238.087

 Change value

− 0.717

− 0.209

− 2.876

4.418

− 2.733

0.717

 CUPT

− 3.733

− 1.089

− 14.977

23.012

− 14.233

3.736

 Trend

Decrease

Decrease

Decrease

Increase

Decrease

Increase

Average precipitation

      

 Original land use

331.042

56.563

77.886

3.441

59.117

159.211

 Grassland transfer to urban land

330.347

56.413

76.034

6.610

57.405

159.955

 Change value

− 0.694

− 0.150

− 1.851

3.169

− 1.711

0.743

 CUPT

− 3.616

− 0.780

− 9.642

16.507

− 8.913

3.872

 Trend

Decrease

Decrease

Decrease

Increase

Decrease

Increase

Low precipitation

      

 Original land use

295.931

46.431

41.031

1.267

25.000

95.041

 Grassland transfer to urban land

295.237

46.313

40.062

3.365

24.154

95.812

 Change value

− 0.694

− 0.119

− 0.969

2.098

− 0.846

0.771

 CUPT

− 3.613

− 0.618

− 5.045

10.925

− 4.405

4.014

 Trend

Decrease

Decrease

Decrease

Increase

Decrease

Increase

Table 8

Response of the main river hydrological indicators in “grassland land transfer to urban land” under the three kinds of precipitation scenarios’ daily scale, monthly scale and annual scale simulations; unit: m3/s

FLOW_IN

High precipitation

Average precipitation

Low precipitation

Original land use

Grassland transfer to urban land

Change value

CUAT

Original land use

Grassland transfer to urban land

Change value

CUAT

Original land use

Grassland transfer to urban land

Change value

CUAT

Daily simulated flood peak

1755

1756

1.000

7.631

1304

1305

1.000

7.631

903.1

903.6

0.500

3.815

Average runoff

            

 January

24.057

24.053

− 0.003

− 0.025

16.587

16.587

0.000

0.000

9.893

9.892

− 0.001

− 0.008

 February

20.470

20.473

0.003

0.025

14.003

14.007

0.003

0.025

8.581

8.581

0.001

0.005

 March

51.787

51.803

0.017

0.127

37.720

37.733

0.013

0.102

24.117

24.123

0.007

0.051

 April

126.133

126.200

0.067

0.509

93.477

93.520

0.043

0.331

62.227

62.263

0.037

0.280

 May

257.733

257.833

0.100

0.763

192.433

192.500

0.067

0.509

127.100

127.167

0.067

0.509

 June

391.167

391.233

0.067

0.509

296.000

296.000

0.000

0.000

199.200

199.267

0.067

0.509

 July

508.967

509.033

0.067

0.509

393.833

393.933

0.100

0.763

264.933

265.033

0.100

0.763

 August

379.167

379.100

− 0.067

− 0.509

293.600

293.567

− 0.033

− 0.254

193.467

193.467

0.000

0.000

 September

507.600

507.600

0.000

0.000

395.467

395.500

0.033

0.254

271.100

271.067

− 0.033

− 0.254

 October

356.300

356.233

− 0.067

− 0.509

278.033

278.000

− 0.033

− 0.254

189.567

189.533

− 0.033

− 0.254

 November

199.867

199.833

− 0.033

− 0.254

154.200

154.133

− 0.067

− 0.509

102.157

102.140

− 0.017

− 0.127

 December

86.210

86.197

− 0.013

− 0.102

61.273

61.267

− 0.007

− 0.051

36.980

36.977

− 0.003

− 0.025

Annual runoff

243.376

243.387

0.011

0.086

186.265

186.275

0.010

0.077

124.575

124.591

0.016

0.122

  1. (a)

    Watershed hydrological analysis

     
The results (Table 7) under the three kinds of precipitation scenarios show: the simulation result’s change trends for ET, SW, PERC, SURQ, GW_Q and WYLD in the “grassland land transfer to urban land” scenario are the same as those in “forest land transfer to urban land”.
  1. (b)

    River hydrological analysis

     

The results (Table 8) under the three kinds of precipitation scenarios show: the simulation result’s change trends for annual runoff and daily flood peak under the “grassland land transfer to urban land” scenario are the same as those under “forest land transfer to urban land”. In addition to individual months, the monthly runoff under the “grassland land transfer to urban land” scenario is largely consistent with that in the “forest land transfer to urban land” scenario.

3.3 Comparative analysis of the two kinds of land use transfer scenarios

Comparing the simulation results for the two types of land transfer scenarios, the CUPT variations of the main watershed hydrological indicators are shown in Fig. 9, while the CUAT variations of the main river hydrological indicators are shown in Fig. 10. The main watershed hydrological indicators from the simulation results under the two types of land transfer scenarios are comparatively analysed: (a) the CUPT variations of ET, SURQ and WYLD, based on annual scale simulation, show “forest land transfer to urban land > grassland transfer to urban land”, while the CUAT variations of daily flood peak and annual runoff show “forest land transfer to urban land > grassland transfer to urban land”. (b) The CUPT of SW, PERC and WYLD, based on annual scale simulation, and the CUAT of simulated monthly runoff do not show a clear trend of change.
Fig. 9

The CUPT variation barplot of main watershed hydrological indicators between the two types of land transfer scenarios

Fig. 10

The CUAT variation barplot of main river hydrological indicators between the two types of land transfer scenarios

  1. (a)

    Watershed hydrological analysis

     
As shown in Fig. 9, the main watershed hydrological indicators’ CUPT variations under the two types of land transfer scenarios are comparatively analysed: (1) the annual scale simulation results for ET, SURQ and WYLD show “forest land transfer to urban land > grassland transfer to urban land”, and the value of the difference between them is 0.342 mm, 0.690 mm and 0.327 mm, respectively, in the high precipitation scenario; 0.441 mm, 0.382 mm, 0.429 mm in the average precipitation scenario; and 0.355 mm, 0.193 mm, 0.312 mm under the low precipitation scenario. (2) The CUPT of SW, PERC and WYLD do not show a clear trend of change.
  1. (b)

    River hydrological analysis

     

As shown in Fig. 10, the main river hydrological indicator’s CUAT variations under the two types of land transfer scenarios are comparatively analysed: (1) the simulated annual runoff shows “forest land transfer to urban land > grassland transfer to urban land”, and the value of the difference between them is 0.028 m3/s in the high precipitation scenario, 0.034 m3/s in the average precipitation scenario and 0.003 m3/s in the low precipitation scenario. (2) The trend of simulated monthly runoff does not show a consistent trend. (3) The change trend of the daily flood peak is the same as that of the simulated annual runoff, and the value of the difference between the two scenarios is 0.471 m3/s, 1.071 m3/s and 3.146 m3/s, respectively, in the high precipitation scenario, average precipitation scenario and low precipitation scenario.

3.4 Discussion

In the analysis of the simulation results, there are two special phenomena.
  1. 1.

    When comparing the CUAT of river hydrological indicator simulated results between the two kinds of land transfer scenarios, the monthly runoff simulation results do not show a consistent trend. The results of the preliminary analysis should be related to the law of growth of trees and grasses, while the effects of different crops on water demand and hydrological cycle in different growth stages are different.

     
  2. 2.

    In the analysis of simulation results for the two kinds of land transfer scenarios, the variation of daily flood peak under the high precipitation scenario is the same as that in the average precipitation scenario based on daily scale simulation. After analysis, it was found that this was caused by the SWAT model’s river channel flow output accuracy.

     

4 Conclusions

Based on the “five sub-basin selection principles” and “two simulation results evaluation indicators” proposed, which can help other scholars carry out similar research, the paper studied the influence of the urbanization process on hydrological processes under different precipitation conditions using the SWAT model. The results were analysed in a multidimensional way which can provide a scientific basis for urban planning and construction for reducing the impact on urban flood.

In the study, we find all the two kinds of land use transfer scenarios show the same laws: (a) with an increase in precipitation, the values of ET, SW, PERC, SURQ, GW_Q and WYLD increase under annual scale watershed simulation. Meanwhile, the simulation results for annual runoff, monthly runoff and daily flood peak also increase. (b) When forest land (or grassland) is transferred to urban land, ET, SW, PERC and GW_Q show a decreasing trend, and the reduction in watershed hydrological indexes is manifested as “high precipitation > average precipitation > low precipitation”. Meanwhile, SURQ, WYLD and annual runoff show an increasing trend, and the increment of SURQ showed “high precipitation > average precipitation > low precipitation”, whereas the increment in WYLD and annual runoff showed “low precipitation > average precipitation > high precipitation”. (c) Analysis of the CUPT of the watershed hydrological indicators shows “SURQ > PERC > GW_Q > ET > SW” in all precipitation scenarios. (d) The change trend of daily flood peak shows “high precipitation = average precipitation > low precipitation” in both kinds of land transfer scenarios.

In addition, compared with grassland, the variations of ET, SURQ and WYLD are greater when forest land transfer to urban land. However, the variations of daily flood peak and annual runoff are just the opposite.

Notes

Acknowledgements

The authors thank the support of the projects of the 13th Five-year Plan of National Scientific Research and Development (Nos. 2017YFC0405803, 2017YFC1502704, 2016YFC0803107, 2016YFC0803109), the IWHR Research & Development Support Program (No. JZ0145B032017) and the National Science Foundation of China (Nos. 41501415, 51420105014). The authors would like to thank National Aeronautics and Space Administration (NASA), Data Center for Resources and Environmental Sciences (RESDC) and Institute of Soil Science (ISSCAS), Chinese Academy of Sciences(CAS) for providing data.

Author contributions

WZ, SH and JL conceived and designed the experiments; WZ and SL performed the experiments; YS and JZ analysed the data; WZ and YF wrote the paper.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

References

  1. Akhter MS, Hewa GA (2016) The use of PCSWMM for assessing the impacts of land use changes on hydrological responses and performance of WSUD in managing the impacts at Myponga Catchment, South Australia. Water 511(8):1–17.  https://doi.org/10.3390/w8110511 CrossRefGoogle Scholar
  2. Arnold JG, Kiniry JR, Srinivasan R, Williams JR, Haney EB, Neitsch SL (2011) Soil & water assessment tool. Input/output documentation, 3rd edn, pp 444–446Google Scholar
  3. Breuer L, Huisman JA, Willems P, Bormann H, Bronstert A, Croke BFW, Frede HG, Gräff T, Hubrechts L, Jakeman AJ, Kite G, Lanini J, Leavesley G, Lettenmaier DP, Lindström G, Seibert J, Sivapalan M, Viney NR (2009) Assessing the impact of land use change on hydrology by ensemble modeling (LUCHEM). I: model intercomparison with current land use. Adv Water Resour 32(2):129–146CrossRefGoogle Scholar
  4. Chen Y, Zhou H, Zhang H, Du GP, Zhou JH (2015) Urban flood risk warning under rapid urbanization. Environ Res 139:3–10.  https://doi.org/10.1016/j.envres.2015.02.028 CrossRefGoogle Scholar
  5. Chu ML, Knouft JH, Ghulam A, Guzman JA, Pan Z (2013) Impacts of urbanization on river flow frequency: a controlled experimental modeling-based evaluation approach. J Hydrol 495:1–12CrossRefGoogle Scholar
  6. Commission European (2017) Directive 2007/60/EC of the European Parliament and of the Council of 23 October 2007 on the assessment and management of flood risks. Official Journal of the European Union 288:27–34Google Scholar
  7. Cornelissen T, Diekkrüger B, Giertz S (2013) A comparison of hydrological models for assessing the impact of land use and climate change on discharge in a tropical catchment. J Hydrol 498(19):221–236CrossRefGoogle Scholar
  8. Dietterick BC, Lynch JA, Corbett ES (1999) A calibration procedure using topmodel to determine suitability for evaluating potential climate change effects on water yield. J Am Water Resour As 35(2):457–468CrossRefGoogle Scholar
  9. Ding YH, Ren GY, Shi GY, Gong P, Zheng XH, Zhai PM, Zhang DE, Zhao ZC, Wang SW, Wang HJ, Luo Y, Chen DL, Gao XJ, Dai XS (2007) National assessment report of climate change(I):climate change in china and its future trend. Adv Clim Change Res 2(1):3–8Google Scholar
  10. Dwarakish GS, Ganasri BP (2015) Impact of land use change on hydrological systems: a review of current modeling approaches. Cogent Geosci 1:1115691.  https://doi.org/10.1080/23312041.2015.1115691 CrossRefGoogle Scholar
  11. Ellis EC, Ramankutty N (2008) Putting people in the map: anthropogenic biomes of the world. Front Ecol Environ 6:439–447.  https://doi.org/10.1890/070062 CrossRefGoogle Scholar
  12. Fletcher TD, Andrieu H, Hamel P (2013) Understanding, management and modelling of urban hydrology and its consequences for receiving waters: a state of the art review. Adv Water Resour 51(1):261–279CrossRefGoogle Scholar
  13. He Y, Lin K, Chen X (2013) Effect of land use and climate change on runoff in the dongjiang basin of south China. Math Probl Eng 2013(1):14–26.  https://doi.org/10.1155/2013/471429 CrossRefGoogle Scholar
  14. Hengade N, Eldho TI (2016) Assessment of LULC and climate change on the hydrology of Ashti Catchment, India using VIC model. J Earth Syst Sci 125(8):1–12CrossRefGoogle Scholar
  15. Hirabayashi Y, Mahendran R, Koirala S, Konoshima L, Yamazaki D, Watanabe S, Kim H, Kanae S (2013) Global flood risk under climate change. Nat Clim Change 3(9):816–821.  https://doi.org/10.1038/nclimate1911 CrossRefGoogle Scholar
  16. Huang SF, Zang WB, Xu M, Li XT, Xie XC, Li ZM, Zhu JS (2015) Study on runoff simulation of the upstream of Minjiang River under future climate change scenarios. Nat Hazards 75(2):139–154CrossRefGoogle Scholar
  17. IPCC (2012) Chapter 3 changes in climate extremes and their impacts on the natural physical environment. In: Managing the risks of extreme events and disasters to advance climate change adaptation.Cambridge University Press, 32 Avenue of the Americas, New York, USA, pp 109–230Google Scholar
  18. Jayakrishnan R, Srinivasan R, Santhi C, Arnold JG (2005) Advances in the application of the SWAT model for water resources management. Hydrol Process 19(3):749–762CrossRefGoogle Scholar
  19. Kaushal SS, Gold AJ, Mayer PM (2017) Land use, climate, and water resources—global stages of interaction. Water 9(815):1–10.  https://doi.org/10.3390/w9100815 CrossRefGoogle Scholar
  20. Lesk C, Rowhani P, Ramankutty N (2016) Influence of extreme weather disasters on global crop production. Nature 529:84–87.  https://doi.org/10.1038/nature16467 CrossRefGoogle Scholar
  21. Liu C, Zhan C, Xia J, Cao J (2014) Review on the influences of climate change and human activities on runoff. J Hydraul Eng 45(4):379–385, 393Google Scholar
  22. Nagelkerke NJD (1991) A note on a general definition of the coefficient of determination. Biometrika 78:691–692.  https://doi.org/10.1093/biomet/78.3.691 CrossRefGoogle Scholar
  23. Nash JE, Sutcliffe JV (1970) River flow forecasting through conceptual models part I—a discussion of principles. J Hydrol 10:282–290.  https://doi.org/10.1016/0022-1694(70)90255-6 CrossRefGoogle Scholar
  24. Neitsch SL, Arnold JG, Kiniry JR, Williams JR (2011) Soil & water assessment tool. Theoretical documentation (Version 2009), pp 3–23Google Scholar
  25. Onstad CA, Jamieson DG (1970) Modeling the effects of land use modifications on runoff. Water Resour Res 6(5):1287–1295CrossRefGoogle Scholar
  26. People’s Daily (2012) Why waterlogging occurred frequency in Chinese big cities. 2012-7-24(4)Google Scholar
  27. Piao SL, Friedlingstein P, Ciais P, Noblet- Ducoudré ND, Labat D, Zaehle S (2007) Changes in climate and land use have a larger direct impact than rising CO2 on global river runoff trends. Proc Natl Acad Sci USA 104(39):15242–15247.  https://doi.org/10.1073/pnas.0707213104 CrossRefGoogle Scholar
  28. Sajikumar N, Remya RS (2015) Impact of land cover and land use change on runoff characteristics. J Environ Manag 161(1):460–468CrossRefGoogle Scholar
  29. Sichuan Institute of Urban Planning & Design (1999) Code for vertical planning on urban field (CJJ 83-99). China Architecture & Building Press, Beijing, China, p 6Google Scholar
  30. Song XM, Zhang JY, Wang GQ, He RM, Wang XJ (2014) Development and challenges of urban hydrology in a changing environment: II: urban stormwater modeling and management. Adv Water Sci 25(5):752–764Google Scholar
  31. Xian LJ, Mu ZX, Jiang HF, Lslom I (2017) Study on impact of climate change on runoff in Pyandj River Basin. Yellow River 39(3):20–23,27Google Scholar
  32. Ye DZ, Fu CB, Ji JJ, Dong WJ (2001) Orderly human activities and subsistence environment. Adv Earth Sci 16(4):453–460Google Scholar
  33. Zhang JY (2012) The vital problem for the urbanization and urban hydrology today. Hydro-Sci Eng 1:1–4Google Scholar
  34. Zhang JY, Song XM, Wang GQ, He RM, Wang XJ (2014) Development and challenges of urban hydrology in a changing environment: I: hydrological response to urbanization. Adv Water Sci 25(4):594–605Google Scholar
  35. Zhao A, Zhu X, Shi P, Pan Y (2013) Review on hydrological response to urbanization at home and abroad. J China Hydrol 33:6–22Google Scholar
  36. Zheng H, Ouyang ZY, Zhao TQ, Li ZX, Xu WH (2003) The impact of human activities on ecosystem services. J Nat Resour 18(1):118–126Google Scholar
  37. Zhou F, Xu Y, Chen Y, Xu CY, Gao Y, Du J (2013) Hydrological response to urbanization at different spatio-temporal scales simulated by coupling of CLUE-S and the SWAT model in the Yangtze River Delta region. J Hydrol 485:113–125CrossRefGoogle Scholar

Copyright information

© Springer Nature B.V. 2019

Authors and Affiliations

  • Wenbin Zang
    • 1
    • 2
  • Shu Liu
    • 2
  • Shifeng Huang
    • 2
    Email author
  • Jiren Li
    • 2
  • Yicheng Fu
    • 1
  • Yayong Sun
    • 2
  • Jingwei Zheng
    • 2
  1. 1.State Key Laboratory of Simulation and Regulation of Water Cycles in River BasinChina Institute of Water Resources and Hydropower ResearchBeijingChina
  2. 2.Research Center on Flood and Drought Disaster Reduction of Ministry of Water ResourcesChina Institute of Water Resources and Hydropower ResearchBeijingChina

Personalised recommendations